Focal segmental glomerular sclerosis (FSGS) is an often progressive form of nephrotic syndrome responsible for 10-15% of end-stage renal disease in children. Patients with FSGS are often resistant to oral steroid therapy, necessitating the use of very agressive alternative treatments (high dose intravenous steroids, cytoxan, plasmapheresis). On the other hand, some patients with FSGS do respond to oral steroids, while a small number of patients with the usually more benign minimal change disease(MCNS) may be resistant to oral steroids. Out study is designed to examine the glomerular hemodynamics, macromolecular permeability and quantitative pathology of pediatric patients with steroid-resistant nephrotic syndrome, in order to try to distinguish at disease onset those patients whose disease will require agresssive therapy from those in whom an extended course of more routine therapy will eventually produce remission. To date, three pediatric patients have been studied: a 15 year old boy and 12 and 15 year old girls. All had FSGS by biopsy. No adults or control nephrotics (with classic relapsing MCNS) have been studied. The analysis of the initial study of the first patient (15 year old girl) is complete and shows a pattern of macromolecular sieving similar to adults with FSGS. This patient has responded to a long course of aggressive therapy and will be approached for a repeat functional study soon. The two other patients were studied within the last 1 1/2 months and their studies have not yet been analyzed. One has already responded rapidly to therapy. Progress has been made on a study of mathematical method of estimation of glomerular volume (Weibel-Gomez method). Preliminary results have been presented at the annual meeting of the American Society of Nephrology (11/97) and will be presented also at the annual GRCR meeting. Briefly, it has been shown using a Monte-Carlo-type simulation analysis that using linear unbiased minimum-variance estimation theory only marginally improves on the efficiency of the classic Weibel-Gomez method, while using censoring (as recommended by many critics of the method) clearly degrades the accuracy of the method.